INTELLIGENT MALWARE DETECTION USING EXTREME LEARNING MACHINE

Main Article Content

V. Keerthi Sree, P. Shravani, V.Sravani, P. Devendar

Abstract

Now-a-days to detect cyber-attack are using static and dynamic analysis of request data. Static analysis is based on signature which we will match existing attack signature with new request packet data to identify packet is normal or contains attack signature. Dynamic analysis will use dynamic execution of program to detect malware/attack, but dynamic analysis is time consuming. To overcome from this problem and to increase detection accuracy with old and new malware attacks, we are using machine learning algorithms and evaluating prediction performance of various machine learning algorithms such as Support Vector Machine (SVM), Random Forest, Decision Tree, Naïve Bayes, Logistic Regression, KNearest Neighbours and Deep Learning Algorithms such as Convolution Neural Networks (CNN) and LSTM (Long Short-Term Memory). Among those, various models Deep learning CNN resulted in superior performance compared to other models. To implement this work and to evaluate machine learning algorithms performance this work using binary malware dataset called ‘MALIMG’. This dataset contains 25 families of malware and application will convert this binary dataset into gray images to generate train and test models for machine learning algorithms. This algorithm converting binary data to images and then generating model, so they are called as MalConv CNN and MalConv LSTM and another algorithm refers as EMBER. Application convert dataset into binary images and then used 80% dataset for training model and 20% dataset for testing. Whenever we upload new test malware binary data then application will apply new test data on train model to predict malware class

Downloads

Download data is not yet available.

Metrics

Metrics Loading ...

Article Details

How to Cite
V. Keerthi Sree, P. Shravani, V.Sravani, P. Devendar. (2023). INTELLIGENT MALWARE DETECTION USING EXTREME LEARNING MACHINE. Turkish Journal of Computer and Mathematics Education (TURCOMAT), 14(2), 50–63. https://doi.org/10.17762/turcomat.v14i2.13522
Section
Research Articles